ing the Theory The theory can be abstracted by recasting key terms more broadly. Social interaction S On the basis of these propositions about joint tasks, the affect theory of exchange offers several hypotheses about structural effects on micro social orders. Aside from analyzing the different forms of exchange (negotiated, etc.), the theory can predict when networks are likely to evolve into groups and when attachments to relations are stronger than to larger social units-that is, networks or groups. (For details, see Lawler 2001a). This content downloaded from 157.55.39.248 on Thu, 28 Jul 2016 05:38:26 UTC All use subject to http://about.jstor.org/terms 12 SOCIAL PSYCHOLOGY QUARTERLY can be substituted for social exchange; joint activities can be substituted for joint tasks; and the activities or tasks involved in an episode of social interaction can vary along the same structural and perceptual dimensions as do exchange tasks-that is, in nonseparability and shared responsibility. The results (benefits) of social interaction produce global emotional responses: successful interaction makes them feel good and unsuccessful interaction makes them feel bad (also see Collins 1981). Under the conditions specified by the affect theory, social-unit attributions for these emotions occur. On the broadest level, the fundamental proposition about micro solidarity is captured by a simple causal chain, portrayed in Figure 2. The upshot is that social interaction involves implicit or explicit joint activities based in a social structure, and these generate benefits for actors. These activities may involve spending a pleasant evening with friends, deciding what cultural event to attend, dividing child care responsibilities in a household, and so forth. Emotional experiences-feeling good or feeling bad-follow from these activities. Although these are not under the actors' direct control, the actors can control them to some extent by understanding the situational conditions under which they occur and by influencing those conditions, or engaging in appropriate approach/avoidance behaviors. These everyday emotions and feelings in social interaction are not only internal reinforcements but also internal signals about the course of social interaction, akin to those analyzed in affect control theory (Heise 1979). According to my theory, they are interpreted in collective terms under specified conditions. Moreover, through the theorized process, successful interaction strengthens the emotional basis of both interpersonal and personto-group ties. Integrating Ideas on Micro Solidarity The idea that repetitive or recurrent interaction is a basis for order, cohesion, or solidarity at the micro level is an underlying theme across a wide variety of theoretical traditions and perspectives. Homans (1961), Berger and Luckmann (1966), Randall Collins (1981), Cook and Emerson (1984), Molm (1994), and Dennis Wrong (1995), for example, incorporate this theme. The relevant literature contains four variations on the theme; these, in turn, represent complementary explanations for the social-order effects of recurrent interaction. One class of explanations is social constructionism. Berger and Luckmann (1966) theorize that "all actions repeated once or more tend to be habitualized to some degree" (p. 57). Actors interpret and ascribe larger meaning to the fact that they interact with the same actors repeatedly; interpretations of habitualization generate "reciprocal typifications"-that is, consensual definitions of the overarching relationship. In this process, the relation or group becomes an object for actors-a "third force" beyond self and other in the social context-and actors orient themselves to this object. Cognitively based expectations of the other are a second type of explanation-that is, the ability to anticipate and predict the other's preferences, beliefs, and behavioral tendencies. Repetitive interaction creates mutual knowledge that enables each actor to form reliable expectations of how the other will behave and respond, and to adjust their behavior in advance. Examples of this approach include the uncertainty-reduction explanation for commitment (Kollock 1994) as well as trust explanations, where trust is defined as an expectation of cooperation (Pruitt and Kimmel 1977; Yamagishi 1995). The overall point is that actors choose to continue interaction with partners with whom they expect mutual cooperation. Social Joint Emotional Micro Structure Activity * Experiences * Solidarity Figure 2. Basic Causal Sequence, Micro Solidarity This content downloaded from 157.55.39.248 on Thu, 28 Jul 2016 05:38:26 UTC All use subject to http://about.jstor.org/terms MICRO SOCIAL ORDERS 13 Normative-based expectations are a third class of explanations for micro social orders. Normative expectations can be imposed exogenously or generated endogenously; in either case, they accord relationships a moral/ethical legitimation. Role theory (Turner 1978) and structural identity theory (Stryker 1980) exemplify this type of explanation. In Stryker's theory, self-other definitions are tied to roles embedded in a social structure, and commitment to those roles depends on the salience of the associated identities. Roles invoke culturally meaningful "names" for actors, and normative expectations are associated with those names. Actors call upon these to guide their own behavior and to anticipate others' behavior, but in the course of social interaction, they also refine or reshape their identities and related expectations. Normative expectations are partly negotiated; thus they are both exogenous and endogenous (see Stryker 1980; Turner 1978). A fourth class of explanations is the emotional/affective. Following Durkheim (1915), Collins's (1981) theory of interaction ritual chains posits explicitly that recurrent social interaction is the most fundamental basis for social order at both macro and micro levels. He ties this idea to the emotional aspects of interaction (i.e., feelings of confidence and uplift). Collins theorizes that social interaction strengthens group solidarity particularly when actors have a common focus of which they are aware and when they experience common emotions or feelings that grow stronger over time (Collins 1981, 1989). Emotional experiences in interaction make group memberships more salient and more real to actors; as a result, actors begin to perceive a moral obligation to one another. The theory of relational cohesion builds on several elements in Collins's theory of interaction